© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 12341
    Prev
    Next
    INDEX
    Explanations

    then return

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-14b-saes/saes_Qwen_Qwen3-14B_batch_top_k/resid_post_layer_20
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æŶéĹ´æ®µ
    -0.21
     Cah
    -0.21
    éĽĨä½ĵç»ıæµİ
    -0.21
    åĩºçĤī
    -0.20
    å°§
    -0.20
     xnxx
    -0.19
    ä¸įåĬ¨äº§
    -0.19
    byn
    -0.19
    综åIJĪå®ŀåĬĽ
    -0.19
    åħ¨è¦ĨçĽĸ
    -0.19
    POSITIVE LOGITS
     foo
    0.24
    Ìģ
    0.23
    .foo
    0.21
    another
    0.20
     crypt
    0.20
     ten
    0.20
    ity
    0.20
    fib
    0.20
     /\
    0.20
    something
    0.20
    Activations Density 1.683%

    No Known Activations